As marketing becomes increasingly data-driven, understanding the fundamentals of machine learning has become essential. This session aims to introduce marketers to the core concepts of machine learning and its application in marketing analytics. As a practical crash course, we will spend the first 1/3 on higher level concepts, the middle 1/3 on pre-processing and model training, and the final 1/3 on inference and presentation.

By the end of this talk, attendees will know how to apply machine learning to several common use cases which include predicting customer churn, identifying the most effective marketing channels, and personalizing content to increase engagement. They will also have gained practical knowledge on how to structure data for machine learning, build accurate models using powerful libraries, and evaluate the outputs of machine learning models.

This talk assumes at least a basic knowledge of statistics (linear regression, statistical tests, etc...), but does not require any software installation or prior coding knowledge.

Takeaway

-Core concepts of machine learning, and how it differs from traditional regression
-Common use cases where applying ML extends existing funnel, attribution, and segmentation analysis.
-Understand how to structure data for machine learning, especially referencing common marketing datasets like event logs, CRM, experiment lists
-How to use powerful ML libraries like XGBoost to get to accurate models with ease
-Know how to evaluate the outputs of machine learning and use explainability tools to show top driving features of predictive models